Pathogenesis of Acute and Chronic Osteomyelitis
Pathogenesis of Acute and Chronic Osteomyelitis
The pathogenetic sequence in development of osteomyelitis might mainly be divided into three major phases: microbial invasion and biofilm proliferation, immune response to bacterial biofilm, and impact of bacterial invasion on bone tissue components. As the first phase reflects a precondition in the pathogenesis of osteomyelitis, the following two phases may occur simultaneously.
Manifestation of osteomyelitis might be dependent on the significance of virulence factors of the microorganisms as well as the patient's immune system or other risk factors, which may have a negative impact on host defence mechanisms like immune deficiency, immunosuppression and systemic diseases (e.g. diabetes mellitus) to name a few examples, but local predisposition factors might also compromise regular bone and wound healing (Table 1).
Staphylococcus aureus, which also predominantly occurs in acute osteomyelitis of children apart from Streptococcus pyogenes and Streptococcus pneumoniae, and coagulase-negative staphylococci species are held responsible in many the cases of chronic osteomyelitis but mycobacteria and fungi have also been detectable. The common feature among these microorganisms comprises the capability to build a three-dimensional matrix or biofilm after colonization of the surfaces of necrotic soft and bone tissues as well as foreign bodies (e.g. osteosynthesis materials, orthopaedic endoprotheses). In the beginning of biofilm formation the microorganisms are in a mainly aerobic and planktonic state with a high metabolic rate, whereas the condition of the metabolism in the fully maturated and sessile form is predominantly anaerobic and extremely reduced.
It is well known that opsonization of planktonic bacteria with immunoglobulin and complement leads to activation of polymorphonuclear neutrophil (PMN)-inducing phagocytosis and generation of reactive oxygen species. Bacterial biofilms are also attacked by PMNs but the immunological efficiency remains dependent on its maturation state. Owing to a comparably lower biofilm clearance, young biofilm structures are easier to target for PMNs. Stroh et al. could show that biofilm opsonization with neither immunoglobulin G nor complement did improve the degree of cellular adherence of PMNs to the biofilm surface, facilitate degranulation or induction of phagocytosis in vitro, whereas the generation of reactive oxygen species was found to critically depend on the biofilm opsonization with immunoglobulin G. Apart from the classical pathways for PMN activation, Meyle et al. were able to identify protein fractions within the nonbacterial-containing extracellular substance of Staphylococcus epidermidis biofilms to induce PMN activity in vitro.
Except for PMNs, the presence of a biofilm leads also to an activation of T-cells and monocytes and resulting altogether in a local increase of proinflammatory cytokines. Owing to so far not completely understood reasons, this activation does not control the focus of infection, but it is assumed that the continuous release of inflammatory mediators is conducting osteolytic and tissue damaging processes.
A significantly higher amount of CD28/CD4 cells at a high level of activation and low proliferation level were found in samples of infected bone when compared with the healthy bone, indicating an increased cytotoxicity by higher expression of CD11b and secretion of perforin compared with CD28 cells. As this characteristic expression was previously found in other chronic disease entities, it was presumed that this cellular pattern might have an impact on osteoclast activity and therefore on bone resorption processes.
Monocytes, which are circulating in the bloodstream, might significantly contribute to bone resorption processes in inflammatory bone disorders. Peripheral blood mononuclear cells (CD14) showed an enhanced rate of cellular adhesion at the vascular endothelium and transendothelial migration, which finally resulted in a comparable higher osteoclast differentiation, when stimulated by tumour necrosis factor (TNF)-[alpha] and interleukin (IL)-1 in vitro.
An increased expression of macrophage inflammatory proteins (MIP1[alpha], CCL3) and MIP2[alpha] (CXCL2) was found in bone samples of patients suffering from infection of orthopaedic prostheses with further close correlation of CD14 as a marker for macrophages and monocytes, respectively. The histological examination also revealed that the macrophage inflammatory proteins were expressed locally and near to monocytic cells. Osteoblasts were also capable to produce macrophage inflammatory proteins in vitro when stimulated by bacteria.
The presence of bacteria also directly influences the cellular components of the bone tissue. In an in-vitro study, it was shown that an infection with S. aureus leads to an increase in expression of Toll-like receptor 2 (TLR2) as a part of the innate immune system, which is known to be upregulated in the process of microbial invasion. Apoptotic cell death was further induced and mitogen-activated protein kinase pathways were activated in osteoblasts. The invasion expression of TLR2 and the activity level of Jun N-terminal kinases (JNK) were directly correlated to have a direct impact on osteoblast apoptosis and ostegenic differentiation after bacterial invasion. It has recently been shown that bacterial endotoxins especially lipopolysaccharide, which localized in the outer membrane of Gram-negative bacteria, was also able to promote apoptosis and inhibit differentiation of osteoblasts by JNK pathway activation.
These findings were confirmed in another in-vitro study, suggesting that methicillin-resistant S. aureus biofilms excrete soluble molecules, which directly impacted the osteoblasts by a decrease in viability and osteogenic potential and indirectly by an increase of the expression of the receptor activator of nuclear factor kappa B (NF-kB) ligand (RANKL) by osteoblasts, which is also known to promote osteoclast activity.
S. aureus also owns another potent virulence factor named protein A (SpA), which is capable to directly bind to osteoblasts and thereby inhibit proliferation and mineralization processes as well as simultaneously induce apoptosis in osteoblasts in vitro. Among others, SpA is able to bind tumour necrosis factor receptor 1 (TNFR-1), which is highly expressed by osteoblasts. This binding activates the NF-kB pathway and results in a release of IL-6, which is commonly known to promote osteoclast activity. In an in-vitro study, perfomed by Claro et al., both a silencing of TNFR-1 in osteoblasts and a destruction of the SpA gene in S. aureus neither led to an activation of NF-kB nor a release of IL-6.
Cytokines belonging to the group of the TNF superfamily are well known to promote apoptotic cell signalling in a high variety of cell types. Here, the soluble TNF-related apoptosis-inducing ligand (TRAIL) may interact with receptors, which are capable to induce apoptosis, for example death receptors and osteoprotegerin (OPG). The latter is believed to act as a decoy soluble receptor for TRAIL and RANKL, which promotes osteoclastogenesis. A study performed by Young et al. showed a constitutive expression of TRAIL receptors in osteoblasts, whereas the death-inducing receptors DR4 as well as DR5 were solely dependent on previous exposure to osteomyelitis-inducing bacteria (S. aureus, Salmonella). It could also be reported that infected osteoblasts showed an inhibition of OPG production. Noninfected osteoblasts did not initiate apoptosis, when TRAIL was administered exogenously but osteoblasts responded after being exposed to bacteria.
Young et al. further hypothesized that an excess of TRAIL expression by osteoblasts would decrease the available OPG leading to an increase of RANKL in the bone tissue promoting osteoclastogenesis. Furthermore, infected osteoblasts would react to an increased presence of TRAIL, which would induce the programmed cell death in osteoblasts.
Osteoblasts are also capable to produce antimicrobial peptides in response to bacterial invasion as a part of the immune system. Human [beta]-defensins represent one main subclass of these antimicrobial peptides and are constitutively increased expressed on mucosal and dermal surfaces to protect the epithelium against microbes. By interaction of these small cationic peptides with the anionic membranes of both Gram-negative and Gram-positive bacteria, defensins, and especially human [beta]-defensin-1 and human [beta]-defensin-2 (hBD-1, hBD-2), are highly efficient to combat a wide range of different pathogens, also further including viral and fungal origins.
Expression of β-defensins in bone tissues has already been demonstrated in healthy and inflamed bone and especially hBD-2 was induced after bacterial contact, leading to the conclusion that antimicrobial peptides might also play a role in osteomyelitis. A decrease in antimicrobial peptide expression was found after administration of immunosuppressive drugs, which was estimated to modulate the susceptibility to osteomyelitis.
In secondary chronic osteomyelitis of the jaws, which mainly occurred after tooth removal and turned into a chronic state, expression of hBD-1 and hBD-2 did not reveal any significant differences, when compared with healthy bone. As infectious agents are capable to induce hBD-2, which was found to be significantly elevated in specimens of bisphosphonate-associated osteonecrosis of the jaws, it was concluded that patients suffering from this specific condition might have a genetic predisposition by lacking an adequate immune response through expression of hBD-2. This assumption was supported by the fact that the osteomyelitis of the jaw after dentoalveolar surgery is a very rare entity.
An increased expression level of hBD-1, hBD-2 and hBD-3 was also found in bone specimens of bisphosphonate-associated osteonecrosis of the jaw, when compared with healthy bone samples and underlining the existing inflammatory components in this bone disorder associated with drugs interfering with the bone metabolism.
Another study further suggested that bisphosphonates even might facilitate bacterial adhesion and consecutively also the bacterial biofilm formation. Different strains of S. aureus and Pseudomonas aeruginosa were cultured on discs consisting of hydroxyapatite coated with pamidronate, in which an increased number of colonies were found compared with control groups. It was concluded that the putative increase in bacterial adhesion was facilitated by bisphosphonates and therefore the development of osteomyelitis might be promoted. In another previously performed study, samples with bisphosphonate-associated osteonecrosis of the jaws were screened for biofilm organization of microorganisms and revealed a coaggregation between different species in the structure of the biofilm consisting of the genus Fusobacterium, Bacillus, Actinomyces, Staphylococcus, Streptococcus, Selenomonas, Treponemes and also Candida species. Furthermore, Lesclous et al. were able to show a statistical link between the amount of inflammatory cells and the clinical appearance of bisphosphonate-associated osteonecrosis, assuming that the bone tissue is affected from an initial site spreading centrifugally even before the necrotic bone is exposed.
Text of Review
The pathogenetic sequence in development of osteomyelitis might mainly be divided into three major phases: microbial invasion and biofilm proliferation, immune response to bacterial biofilm, and impact of bacterial invasion on bone tissue components. As the first phase reflects a precondition in the pathogenesis of osteomyelitis, the following two phases may occur simultaneously.
Microbial Invasion and Biofilm Proliferation
Manifestation of osteomyelitis might be dependent on the significance of virulence factors of the microorganisms as well as the patient's immune system or other risk factors, which may have a negative impact on host defence mechanisms like immune deficiency, immunosuppression and systemic diseases (e.g. diabetes mellitus) to name a few examples, but local predisposition factors might also compromise regular bone and wound healing (Table 1).
Staphylococcus aureus, which also predominantly occurs in acute osteomyelitis of children apart from Streptococcus pyogenes and Streptococcus pneumoniae, and coagulase-negative staphylococci species are held responsible in many the cases of chronic osteomyelitis but mycobacteria and fungi have also been detectable. The common feature among these microorganisms comprises the capability to build a three-dimensional matrix or biofilm after colonization of the surfaces of necrotic soft and bone tissues as well as foreign bodies (e.g. osteosynthesis materials, orthopaedic endoprotheses). In the beginning of biofilm formation the microorganisms are in a mainly aerobic and planktonic state with a high metabolic rate, whereas the condition of the metabolism in the fully maturated and sessile form is predominantly anaerobic and extremely reduced.
Immune Response to Bacterial Biofilm
It is well known that opsonization of planktonic bacteria with immunoglobulin and complement leads to activation of polymorphonuclear neutrophil (PMN)-inducing phagocytosis and generation of reactive oxygen species. Bacterial biofilms are also attacked by PMNs but the immunological efficiency remains dependent on its maturation state. Owing to a comparably lower biofilm clearance, young biofilm structures are easier to target for PMNs. Stroh et al. could show that biofilm opsonization with neither immunoglobulin G nor complement did improve the degree of cellular adherence of PMNs to the biofilm surface, facilitate degranulation or induction of phagocytosis in vitro, whereas the generation of reactive oxygen species was found to critically depend on the biofilm opsonization with immunoglobulin G. Apart from the classical pathways for PMN activation, Meyle et al. were able to identify protein fractions within the nonbacterial-containing extracellular substance of Staphylococcus epidermidis biofilms to induce PMN activity in vitro.
Except for PMNs, the presence of a biofilm leads also to an activation of T-cells and monocytes and resulting altogether in a local increase of proinflammatory cytokines. Owing to so far not completely understood reasons, this activation does not control the focus of infection, but it is assumed that the continuous release of inflammatory mediators is conducting osteolytic and tissue damaging processes.
A significantly higher amount of CD28/CD4 cells at a high level of activation and low proliferation level were found in samples of infected bone when compared with the healthy bone, indicating an increased cytotoxicity by higher expression of CD11b and secretion of perforin compared with CD28 cells. As this characteristic expression was previously found in other chronic disease entities, it was presumed that this cellular pattern might have an impact on osteoclast activity and therefore on bone resorption processes.
Monocytes, which are circulating in the bloodstream, might significantly contribute to bone resorption processes in inflammatory bone disorders. Peripheral blood mononuclear cells (CD14) showed an enhanced rate of cellular adhesion at the vascular endothelium and transendothelial migration, which finally resulted in a comparable higher osteoclast differentiation, when stimulated by tumour necrosis factor (TNF)-[alpha] and interleukin (IL)-1 in vitro.
An increased expression of macrophage inflammatory proteins (MIP1[alpha], CCL3) and MIP2[alpha] (CXCL2) was found in bone samples of patients suffering from infection of orthopaedic prostheses with further close correlation of CD14 as a marker for macrophages and monocytes, respectively. The histological examination also revealed that the macrophage inflammatory proteins were expressed locally and near to monocytic cells. Osteoblasts were also capable to produce macrophage inflammatory proteins in vitro when stimulated by bacteria.
Impact of Bacterial Invasion on Bone Tissue Components
The presence of bacteria also directly influences the cellular components of the bone tissue. In an in-vitro study, it was shown that an infection with S. aureus leads to an increase in expression of Toll-like receptor 2 (TLR2) as a part of the innate immune system, which is known to be upregulated in the process of microbial invasion. Apoptotic cell death was further induced and mitogen-activated protein kinase pathways were activated in osteoblasts. The invasion expression of TLR2 and the activity level of Jun N-terminal kinases (JNK) were directly correlated to have a direct impact on osteoblast apoptosis and ostegenic differentiation after bacterial invasion. It has recently been shown that bacterial endotoxins especially lipopolysaccharide, which localized in the outer membrane of Gram-negative bacteria, was also able to promote apoptosis and inhibit differentiation of osteoblasts by JNK pathway activation.
These findings were confirmed in another in-vitro study, suggesting that methicillin-resistant S. aureus biofilms excrete soluble molecules, which directly impacted the osteoblasts by a decrease in viability and osteogenic potential and indirectly by an increase of the expression of the receptor activator of nuclear factor kappa B (NF-kB) ligand (RANKL) by osteoblasts, which is also known to promote osteoclast activity.
S. aureus also owns another potent virulence factor named protein A (SpA), which is capable to directly bind to osteoblasts and thereby inhibit proliferation and mineralization processes as well as simultaneously induce apoptosis in osteoblasts in vitro. Among others, SpA is able to bind tumour necrosis factor receptor 1 (TNFR-1), which is highly expressed by osteoblasts. This binding activates the NF-kB pathway and results in a release of IL-6, which is commonly known to promote osteoclast activity. In an in-vitro study, perfomed by Claro et al., both a silencing of TNFR-1 in osteoblasts and a destruction of the SpA gene in S. aureus neither led to an activation of NF-kB nor a release of IL-6.
Cytokines belonging to the group of the TNF superfamily are well known to promote apoptotic cell signalling in a high variety of cell types. Here, the soluble TNF-related apoptosis-inducing ligand (TRAIL) may interact with receptors, which are capable to induce apoptosis, for example death receptors and osteoprotegerin (OPG). The latter is believed to act as a decoy soluble receptor for TRAIL and RANKL, which promotes osteoclastogenesis. A study performed by Young et al. showed a constitutive expression of TRAIL receptors in osteoblasts, whereas the death-inducing receptors DR4 as well as DR5 were solely dependent on previous exposure to osteomyelitis-inducing bacteria (S. aureus, Salmonella). It could also be reported that infected osteoblasts showed an inhibition of OPG production. Noninfected osteoblasts did not initiate apoptosis, when TRAIL was administered exogenously but osteoblasts responded after being exposed to bacteria.
Young et al. further hypothesized that an excess of TRAIL expression by osteoblasts would decrease the available OPG leading to an increase of RANKL in the bone tissue promoting osteoclastogenesis. Furthermore, infected osteoblasts would react to an increased presence of TRAIL, which would induce the programmed cell death in osteoblasts.
Osteoblasts are also capable to produce antimicrobial peptides in response to bacterial invasion as a part of the immune system. Human [beta]-defensins represent one main subclass of these antimicrobial peptides and are constitutively increased expressed on mucosal and dermal surfaces to protect the epithelium against microbes. By interaction of these small cationic peptides with the anionic membranes of both Gram-negative and Gram-positive bacteria, defensins, and especially human [beta]-defensin-1 and human [beta]-defensin-2 (hBD-1, hBD-2), are highly efficient to combat a wide range of different pathogens, also further including viral and fungal origins.
Expression of β-defensins in bone tissues has already been demonstrated in healthy and inflamed bone and especially hBD-2 was induced after bacterial contact, leading to the conclusion that antimicrobial peptides might also play a role in osteomyelitis. A decrease in antimicrobial peptide expression was found after administration of immunosuppressive drugs, which was estimated to modulate the susceptibility to osteomyelitis.
In secondary chronic osteomyelitis of the jaws, which mainly occurred after tooth removal and turned into a chronic state, expression of hBD-1 and hBD-2 did not reveal any significant differences, when compared with healthy bone. As infectious agents are capable to induce hBD-2, which was found to be significantly elevated in specimens of bisphosphonate-associated osteonecrosis of the jaws, it was concluded that patients suffering from this specific condition might have a genetic predisposition by lacking an adequate immune response through expression of hBD-2. This assumption was supported by the fact that the osteomyelitis of the jaw after dentoalveolar surgery is a very rare entity.
An increased expression level of hBD-1, hBD-2 and hBD-3 was also found in bone specimens of bisphosphonate-associated osteonecrosis of the jaw, when compared with healthy bone samples and underlining the existing inflammatory components in this bone disorder associated with drugs interfering with the bone metabolism.
Another study further suggested that bisphosphonates even might facilitate bacterial adhesion and consecutively also the bacterial biofilm formation. Different strains of S. aureus and Pseudomonas aeruginosa were cultured on discs consisting of hydroxyapatite coated with pamidronate, in which an increased number of colonies were found compared with control groups. It was concluded that the putative increase in bacterial adhesion was facilitated by bisphosphonates and therefore the development of osteomyelitis might be promoted. In another previously performed study, samples with bisphosphonate-associated osteonecrosis of the jaws were screened for biofilm organization of microorganisms and revealed a coaggregation between different species in the structure of the biofilm consisting of the genus Fusobacterium, Bacillus, Actinomyces, Staphylococcus, Streptococcus, Selenomonas, Treponemes and also Candida species. Furthermore, Lesclous et al. were able to show a statistical link between the amount of inflammatory cells and the clinical appearance of bisphosphonate-associated osteonecrosis, assuming that the bone tissue is affected from an initial site spreading centrifugally even before the necrotic bone is exposed.